JOVANA
Explore Library Glossary Getting Started Three Levels Fields How it works Mission
Join the mission
All guides

Hardenability and the Jominy Test

Guide 3 taught you to quench a steel hard and temper it tough. But quench a thick bar and only the skin turns hard — the core stays soft. Hardenability is the depth of hardening, a different property from hardness, and the Jominy end-quench test measures it on a single clever specimen.

Hardness is not hardenability

In guide 3 you learned the recipe that shocks steel glass-hard and then relaxes it just enough to be tough: austenitize, quench to martensite, then temper. That recipe hides a problem the moment your part is more than paper-thin. When you plunge a bar into water, only the surface actually cools fast enough to become martensite. Heat buried in the middle has to crawl outward through the steel before it can escape, so the core cools slowly — and slow cooling makes soft pearlite, not hard martensite. A quenched bar can therefore be file-hard on the outside and still bend like a soft rod at its heart.

This forces us to split one word into two. Hardness, which you met back in the mechanical-properties rung, is simply how hard a spot on the surface ends up — how much it resists a pressed indenter, read off a Rockwell or Brinell scale. Hardenability is a completely different question: how deep below the surface does the hard martensitic structure reach? One is a number at a point; the other is a depth profile. A shallow-hardening steel and a deep-hardening steel can reach the very same peak surface hardness and yet behave nothing alike inside a thick section.

Why the core stays soft: a race against the CCT clock

To see what really controls the depth of hardening, go back to the continuous-cooling-transformation diagram from the transformations rung. Austenite that is cooling has a deadline: if it dawdles it will begin turning into soft pearlite or medium bainite at a certain 'nose' of time and temperature. Martensite only forms when the steel dives past that nose so fast that austenite never gets a chance to transform the ordinary way — the carbon simply has no time to diffuse out, and the lattice snaps into strained, carbon-trapped martensite instead.

Now picture the cooling rates inside one quenched bar. The surface, in direct contact with the water, plunges fastest — it clears the nose and turns to martensite. A millimetre in, the cooling is a touch slower; deeper still, slower again; and dead centre, the heat has the furthest to travel and cools slowest of all. Somewhere along that gradient a critical depth is reached where the local cooling curve just grazes the nose of the CCT diagram. Outside that depth: martensite. Inside it: the cooling curve clips the nose, transformation begins, and you get pearlite and bainite. Hardenability is nothing more mysterious than how far in that critical depth sits — and that, in turn, is set by how far to the right the CCT nose has been pushed.

This is the whole reason alloy steels exist. Dissolve chromium, molybdenum, nickel, manganese or a trace of boron into the austenite and they slow down the diffusion that pearlite needs to form, sliding the entire CCT nose to the right — buying more time. With more time, even the slowly-cooled core can still miss the nose and turn martensitic. A plain-carbon steel has its nose so far left that only a thin skin ever beats it; a well-alloyed steel can through-harden a bar the thickness of your wrist. Carbon still fixes the peak hardness; the alloying elements are there almost entirely to buy hardenability.

The Jominy end-quench test

How do you put a number on 'depth of hardening'? The clever trick is the Jominy end-quench test, and its beauty is that one small specimen samples every cooling rate at once. You machine a standard round bar (25 mm across, 100 mm long), heat it until it is fully austenite, then stand it up and squirt a jet of room-temperature water at its bottom face only. The quenched end cools ferociously fast; the far end, cooled only by air, cools slowly; and every point between them cools at a smoothly-decreasing rate. A single bar thus reproduces the whole spread of cooling rates you would find from the surface to the core of many differently-sized parts.

  1. Austenitize the standard 25 mm x 100 mm bar: hold it above the transformation temperature until it is uniformly austenite (typically ~30 min).
  2. Drop it into the fixture and hit only the bottom face with a controlled jet of water at 24 degrees C. The far end never touches water — it just loses heat to the air.
  3. After it has fully cooled, grind two shallow flats along the length, then measure Rockwell C hardness at fixed intervals (every 1.6 mm) marching away from the quenched end.
  4. Plot hardness against distance from the quenched end. That curve is the steel's hardenability signature: a flat, high plateau means deep hardening; a steep early drop means shallow hardening.
Jominy end-quench curves -- Rockwell C hardness vs distance from quenched end

 HRC
  60 +oo                        cooling rate: FAST <----------- SLOW
     |  ooo   4340 (Ni-Cr-Mo low-alloy)
  50 +     ooooo
     |          ooooooooo
  40 +    xx             ooooooooooooooooooooooo   <- stays hard, DEEP
     |      x
  30 +       xx   1040 (plain carbon)
     |         xxxx
  20 +            xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx   <- falls off, SHALLOW
     +--+----+----+----+----+----+----+----+----+--> distance from
        0    5   10   15   20   25   30   40   50    quenched end (mm)
     martensite  ---- cooling rate drops ---->  pearlite / bainite

 Same carbon (~0.40%) => same peak hardness at the quenched end (~55 HRC).
 Different alloying   => wildly different DEPTH of hardening.
Two 0.40% carbon steels tell the whole story. Both reach roughly the same peak hardness at the water-quenched end (carbon sets the peak). But plain-carbon 1040 tumbles within about 10 mm, while low-alloy 4340 holds its hardness out to 50 mm — the alloying pushed the CCT nose right, so martensite forms even where cooling is slow.

What you tune: alloy, grain, and the quench itself

Three levers set where the Jominy curve lands. The first is alloy content, which we have met — chromium, molybdenum, manganese, nickel and boron all push the nose right and deepen hardening, which is exactly what separates a low-alloy steel from a plain-carbon one. The second is prior austenite grain size, and here honesty matters: coarser austenite grains actually raise hardenability, because grain boundaries are the favourite nucleation sites for pearlite, so fewer boundaries means slower transformation and a nose shifted right. But you almost never want coarse grains — they wreck toughness and invite brittle fracture. So this is a genuine trade-off, not a free lunch: the grain size that is best for depth of hardening is the worst for impact resistance.

The third lever is the quench medium — how fiercely you pull the heat out. A brine or water quench is severe; oil is gentler; still air is mild, and agitation makes any of them harsher. Faster extraction hardens deeper, so it is tempting to always quench in water. But there is a price, and it is thermal stress: when the surface contracts violently while the core is still hot and expanded, the mismatch can warp the part or split it outright in a quench crack. This is the elegant payoff of hardenability. A steel with high hardenability can be through-hardened in gentle oil instead of savage water, so you get the hard structure you need and far less distortion and cracking. You buy that forgiveness with alloy content up front.

Case hardening: a hard skin over a tough heart

Sometimes shallow hardening is not a flaw to fight but exactly what you want. Think of a gear tooth or a camshaft: the surface must be glass-hard to resist wear and scuffing, yet the body must stay tough to absorb shock without snapping. A fully-hardened part is too brittle for that; a fully-soft one wears out. The answer is case hardening — deliberately engineering a hard outer case over a soft, tough core, so a single part gets both personalities at once.

The most common route is carburizing, and it is a lovely direct use of diffusion. Pack a low-carbon steel part in a carbon-rich atmosphere and hold it hot in the austenite range; carbon atoms dissolve into the surface and, by diffusion, soak inward — steeply concentrated at the skin, tailing off with depth exactly as Fick's laws predict. Now quench. The carbon-enriched skin has enough carbon to transform into hard martensite, while the low-carbon core, poor in carbon, stays soft and ductile even at the same cooling rate. You have, in effect, drawn a hardenability gradient right into a single part by changing its chemistry with depth rather than its cooling rate.

There is a bonus that engineers prize even more than the hardness. Because martensite is less dense than the austenite it forms from, the surface skin tries to expand as it transforms — but the core has already set, and it clamps the skin, leaving the surface locked in compression. That built-in compressive layer is a gift against fatigue: fatigue cracks are born at the surface under tension, and a surface already squeezed in compression is far more reluctant to crack. This is why carburized gears and shafts routinely outlive their through-hardened cousins in service — the honest reason case hardening dominates high-cycle, high-wear parts.